Echo-DM: Ultrasound Marker Removal via Conditional Latent Diffusion and Region-Aware Fusion
2026-06-08 • Computer Vision and Pattern Recognition
Computer Vision and Pattern Recognition
AI summaryⓘ
The authors address the problem of unwanted markers like measurement calipers on ultrasound images, which can confuse AI models analyzing these images. They propose Echo-DM, a method that removes these markers without needing extra guidance like masks, while keeping important image details intact. Their approach uses a special neural network design with diffusion and fusion steps to clean the images effectively. Tests on a large dataset show their method works better than previous ones, balancing quality and speed. They also provide different versions of their method to suit various needs.
ultrasound imagingartificial markersshortcut biasdeep learninglatent diffusionregion-aware fusionencoder-decoder networkVAE (Variational Autoencoder)RAE (Residual Autoencoder)image restoration
Authors
Zhiwei Wang, Tao Huang, Wentao Jiang, Muyi Li, Jianxin Liu, Jian Chen, Jie Zou, Yong Luo, Bo Du, Jing Zhang
Abstract
Clinical ultrasound images often contain artificial markers, such as measurement calipers and text, to assist diagnostic interpretation and comparison. However, these markers can introduce shortcut bias in downstream automated analysis, encouraging deep learning models to rely on marker-related cues rather than clinically meaningful anatomy. Existing marker removal methods are either mask-dependent and vulnerable to error propagation, or mask-free deterministic restorers that may over-smooth ultrasound texture and perturb unaffected background regions. To address these challenges, we present Echo-DM, a framework for ultrasound marker removal via conditional latent diffusion and region-aware fusion. Echo-DM follows a common encoder-diffusion-decoder pipeline, where a DiT-based conditional latent diffusion network performs global restoration and a region-aware fusion module enforces preservation-aware image-space refinement under end-to-end mask-free inference. Building on this fixed core design, we further instantiate Echo-DM-V and Echo-DM-R with VAE-based and RAE-based latent modules, respectively, which demonstrates that the Echo-DM architecture is compatible with diverse latent-module instantiations. Extensive experiments on Echo-PAIR, a large-scale paired clinical ultrasound dataset, demonstrate superior marker removal and strong anatomical fidelity compared with representative two-stage baselines, while providing favorable quality--efficiency trade-offs across deployment settings. Data, code and models will be released at https://github.com/MiliLab/Echo-DM.